Benchgen

SWE-bench Pro — Results

RankModelScore
1flint89.7
2fugu-ultra73.7
3qwen3-8-max67.7
4beam-1-0-501b-a23b66.1
5hy4-preview65.7
6nex-n2-5-max65.7
7ornith-1-5-397b65.1
8gpt-5-6-sol64.6
9claude-opus-4-764.3
10gpt-5-6-terra63.4
11claude-sonnet-563.2
12gpt-5-6-luna62.7
13qwen3-8-flash-next62.5
14glm-5-262.1
15qwen3-8-27b61.7
16nex-n2-5-pro61.2
17sage-a1-561.2
18dots3-note-preview61
19nanbeige4-3-33b60.3
20ornith-1-5-35b-a3b59.6
21laguna-s-2-159.4
22fugu59
23gpt-5-558.6
24kimi-k2-658.6
25claude-sonnet-4-658.1

SWE-bench Pro — Evaluations

1 phaseActive

No evaluations yet for this benchmark.